Pulmonary surfactant is a surface active material composed of a lipids/protein mixture that forms layered structures, i.e. Langmuir monolayer at the alveolar air/water interface and bilayers in lamellar bodies. The typical compositions of pulmonary surfactant are 50% of dipalmitolphosphotidylcholine (DPPC), 25% unsaturated phosphotidylcholine, 12% of the negatively charged phosphatidylglycerol (PG), and 8% of surfactant proteins. The remaining 5% includes other type of lipids and cholesterol. Various pathological conditions have been associated with abnormal composition levels of pulmonary surfactants. For example, an insufficient amount of pulmonary surfactants in premature or newly born infants causes respiratory distress syndrome, which is a major cause of neonatal mortality. Large quantities of certain surfactant proteins have been associated to the formation of amyloidal fibrils causing a condition termed pulmonary alveolar proteinosis (PAP). Also, inappropriate concentration of lipids in the lung surfactant can induce lipid aggregation, causing degenerative diseases such as acute lung injury and lipidosis. The detailed molecular mechanism for the aggregation of the various components of lung surfactant is not fully understood. We hypothesize that aggregation of surfactant proteins and/or lipids will occur at certain lipid layered-structure and thermodynamic states (composition, temperature and surface tension). The main reason for our hypothesis is that intermolecular forces in the lipidic-layered environment can promote the necessary structural changes in proteins, and hence the formation of aggregates. We also hypothesize that general rules can be derived allowing us to make predictions regarding the formation of aggregates in lung surfactants. To investigate our hypotheses, we propose to use state-of-the-art computer simulations. Specifically, coarse-grained, united, and all-atom interparticle potentials will be used and/or developed that permit an accurate description of the forces acting on the systems being investigated. Computational techniques such as Monte Carlo and Molecular Dynamics, coupled to efficient sampling techniques, will be used to determine the properties and conditions for the formation of aggregates in pulmonary surfactants. The research will permit the elucidation of the mechanism of formation of lipid/protein aggregates in lung surfactants and the correlation of this formation to pathologies. This correlation will help to the development of new approaches for the treatment of PAP and lipidosis, as well as other protein aggregation related diseases.